Recent advances in neurophysiology have allowed researchers to study the activity of groups of neurons with high temporal resolution and in specific locations in the brain. These advances create the possibility for brain-machine interfaces allowing an amputee to control a prosthetic limb in much the same way that person would control a natural limb. Although noninvasive sensors, such as multichannel electroencephalogram (EEG), have shown some promise as simple interfaces to computers, they do not currently offer the spatial resolution needed for prosthetic control. Current research into the electrical activity of small groups of neurons has thus been done primarily with arrays of microelectrodes inserted into the brain.
Current intra-cortical microelectrode recording systems can record electrical signals from groups of neurons. These systems typically use a microscopic tapered conductive element, insulated except at its tip, to record the neuron signals. Other conductor designs, such as blunt cut wires, may record single neurons, but have sub-optimal recording characteristics. Further, nearly all recording systems rely on arrays of fixed electrodes connected to data acquisition systems through long wiring or cable harnesses. The percutaneous connectors associated with these cables present a potential source of infection that limits the useful life of these systems. The cables themselves also present additional problems in the design of a prosthesis that must continue to function over many years and not interfere with the patient's daily life. For instance, the cables limit the patient's mobility by being tethered to a signal processing device. Relatively long cables may also present a source of electrical interference and may break after repetitive use.
The current microelectrode systems for recording single neurons can be grouped into two broad classes: those having microdrive mechanisms and those having fixed electrode arrays. Systems with microdrive mechanisms allow one to vertically position the electrodes in the brain tissue. Thus, a user can actively search for neurons of interest and accurately position the electrode tip near the soma of the neuron to improve the signal-to-noise ratio. These systems, however, have their disadvantages. First, even individual microdrive systems are bulky and cannot be fully implanted in a human. Second, microdrive systems typically cannot use more than a few dozen electrodes due to space limitations and the time it takes to independently position each electrode near a neuron.
Fixed electrode array systems overcome some of these problems, but have their own problems as well. Once placed in the brain, fixed electrode arrays can not be repositioned, so they rely on chance proximity to neurons. The most basic fixed electrode arrays record neural activity using multiple micro-wires or hatpin-like electrodes individually inserted into the brain. Because it can take a relatively significant amount of time to insert each electrode, however, these systems have not been widely used. More recently, wire bundles have been developed which are inserted into the cortex as a unit, but they lack features of ideal recording electrodes, such as tip shape, overall size, and impedance. In particular, the common square tip of such microwires can damage the cortex and can have difficulty penetrating the tough cerebral membranes, as well as brain tissue.
A major disadvantage of these fixed array systems is that they do not offer the ability to actively hunt for neurons since the electrode tips cannot be easily placed near the soma of the neurons. To help overcome this, large numbers of electrodes are inserted to increase the chance that the electrodes are positioned in close proximity to neurons. The input impedances of the electrodes may also be lowered to enhance their ability to record distant signals. Lowering the input impedance, however, also lowers the signal-to-noise ratio.
Accordingly, there is a need for a fixed microelectrode array system that may have numerous electrodes providing a high signal-to-noise ratio. Further, there is a need for a fixed array system that has a flexible design and that does not rely upon percutaneous cabling systems to communicate with a data acquisition system.